,
10 tweets,
2 min read
Read on Twitter
I find this interesting, but only likely to work in newer subdivisions where the housing is reasonably identical and maintenance differentials don't matter too much.
In my neighborhood, for example, most of the homes were built between 1885 and 1910. Massive differences in price both block by block (my block has a beautiful arch of 100 year old trees and is fairly quiet for the area; one block south has no trees and gets more traffic noise.
The interiors of the houses vary widely, from renovated-two-months-ago-with-the-latest-in-builder-grade-white-kitchens to holes-in-the-ceiling-and-hey-when's-the-last-time-you-saw-knob-and-tube-wiring?
The prices for such homes can vary by nearly 100%, with all sorts of intermediate grades, like "decently renovated in the 1980s but everything's creaky" to "badly renovated in the 1990s" to "renovated in 2005, and dear God, what were they thinking with that layout?"
I just don't see an algorithm figuring that out.
The bigger problem is that the algorithms inherently rely on the traditional market to get their prices. If they drive the market out of business, that data goes away.
The bigger problem is that the algorithms inherently rely on the traditional market to get their prices. If they drive the market out of business, that data goes away.
Now, to be fair, that could be a profit opportunity! A nice, consolidated market with a middleman doing the buying and selling could open up big profit opportunities for the middleman, underpaying to buy homes, overcharging to sell them.
But I think that homes are just too big a portion of net worth, particularly in cities, for people to allow a middleman to take that kind of profit. In most transactions, they're going to have to buy within a percent of the market, and sell there too.
Effectively, their profit margin should be locked at close to the local agent commission. They'll get a small premium for convenience, but in order to get that premium, they'll have to assume the pricing and time-to-sell risk, making this at best, ultimately a wash.
How big is the pool of capital that wants to take on illiquid assets for a max 5-7% profit?
To be clear, I think there's some market for this, with low-information buyers who tend to be hyperbolic discounters. Those with whom the "we buy homes for CASH!!!" flippers trade.
To be clear, I think there's some market for this, with low-information buyers who tend to be hyperbolic discounters. Those with whom the "we buy homes for CASH!!!" flippers trade.
That's just not most of the housing market.
